Accelerate your data science workflow from months to days with foundation models for tabular data.
Project description
💠 Future Frame
- This Python package allows you to interact with pre-trained foundation models for tabular data.
- Easily fine-tune them on your classification and regression use cases in a single line of code.
- Interested in what we're building? Join our waitlist.
Installation
Install Future Frame with pip
– more details on our PyPI page.
pip install futureframe
Quick Start
Use Future Frame to fine-tune a pre-trained foundation model on a classification task.
# Import standard libraries
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.metrics import roc_auc_score
# Import Future Frame
import futureframe as ff
# Import data
dataset_name = "tests/data/churn.csv"
target_variable = "Churn"
df = pd.read_csv(dataset_name)
# Split data
X, y = df.drop(columns=[target_variable]), df[target_variable]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
# Fine-tune a pre-trained classifier with Future Frame
model = ff.models.CM2Classifier()
model.finetune(X_train, y_train)
# Make predictions with Future Frame
y_pred = model.predict(X_test)
# Evaluate your model
auc = roc_auc_score(y_test, y_pred)
print(f"AUC: {auc:0.2f}")
Models
Model Name | Paper Title | Paper | GitHub |
---|---|---|---|
CM2 | Towards Cross-Table Masked Pretraining for Web Data Mining | Ye et al., 2024 | Link |
More foundation models will be integrated into the library soon. Stay stuned by joining our waitlist!
Links
- Future Frame Official Website
futureframe
PyPI Pagefutureframe
GitHub Repository- Documentation: coming soon!
Contributing
- We are currently under heavy development.
- If you'd like to contribute, please send us an email at eduardo(at)futureframe.ai.
- To report a bug, please write an issue.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
futureframe-0.1.5.tar.gz
(29.8 kB
view hashes)
Built Distribution
Close
Hashes for futureframe-0.1.5-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 25a2347bbbaff9ca54556d55a4f731e0a717ba605af59e0c893278a5c17c9289 |
|
MD5 | 60a7cf31587a96d2fa40b0cbda1ea540 |
|
BLAKE2b-256 | c604787fbf01a1705aa6c39115a346ddc16a72dfbcfab2fab80001a0a832891e |